Through epidemiological research, a consensus has been reached that environmental factors are important in the etiology of breast cancer. Genetic analysis of breast cancer pedigrees has proven that hereditary factors can also contribute to breast cancer etiology. The guiding principal of this project and its two interactive R01 partners is to discover the environmental and hereditary factors that define an individual woman's breast cancer risk. This proposal concentrates on two issues important to the genetic epidemiology of breast cancer. 1) Genes that contribute to breast cancer risk need to be discovered. Despite recent exciting advances in breast cancer genetics, the discovery of BRCA1 located on 17q, there is compelling evidence that there are additional genes to be discovered. In this proposal, we will use a population based sampling of breast cancer affected relatives to perform linkage studies. We will use APM linkage strategies because they are free of the assumptions that are required to perform likelihood linkage analysis. 2) Once breast cancer susceptibility genes are found, they must then be incorporated into epidemiological studies: efficient assays need to be developed; the population genetic characteristics need to be determined; and large numbers of individuals need to be assayed if genetic/environmental interactions are to be discovered. We will sequence cases and controls at the p53 gene concentrating on those exons that have not been analyzed by other researchers. We will also perform sequence analysis of BRCA1 if it is isolated during this granting period. Interpretation of the data will use case-control methodologies and will concentrate on the potential of finding gene/environment interactions. This proposal represents the next generation of genetic analysis. We will be searching for the genes that are more difficult to find; genes that do not present themselves in large, highly penetrant pedigrees; genes that may be more relevant to the total population incidence of breast cancer. The limitation we face at this time is not genotyping technology, but rather patient populations and analysis methods that are sufficient to the task. This requires the merging of genetics, epidemiology and biostatistics. The foundation of this proposal, and the other two interactive projects, is that we have already ascertained 1400 women who were diagnosed with breast cancer before age 45 and a similar number of matched controls. Blood samples have been collected from over half the cases and 200 controls. A relatively small amount of additional effort is required to fully transform this epidemiological resource into the resource we need for our proposed genetic studies. Application and/or development of appropriate biostatistical tools requires interaction with our partner projects.